What's new
What's new

Vintage VFD?

NYShaper

Plastic
Joined
Mar 31, 2019
I bought this magic box from a retiring machinist who said that it is some sort of early VFD that he had built by an automation control company in Allentown PA in the 1980's or 1990's to run a 7.5HP 16x60 lathe on single phase power. I talked to the gentleman for over an hour and he seemed to be a straight shooter but didn't remember many details about the unit. We used it to power up an unloaded 3HP 3 phase motor at his shop and verified that the speed control did adjust the motor rpms. He wanted less than the cost of a bottle of whisky for it, so I took a chance.
The components seem kind of similar to the early 1980's VFD pictured here but I cant identify enough similarities to be sure what it is?
What is a Variable Frequency Drive?

The unit appears to be really well made and the heat sink that makes up the entire bottom of the unit looks sized for a high amperage application, but the other components don't seem like they are sized for a 7.5HP load. Does anybody have any thoughts on what this could be or how to determine its capacity?
UNADJUSTEDNONRAW_thumb_d846.jpgUNADJUSTEDNONRAW_thumb_d845.jpgUNADJUSTEDNONRAW_thumb_d844.jpgUNADJUSTEDNONRAW_thumb_d881.jpgUNADJUSTEDNONRAW_thumb_d887.jpg
 
Well, it surely is an old one.

The TO-3 and TO-66 transistors are bipolar transistors, used in quantity because of technical issues that such transistors had at the time, as well as the high currents and voltages.

It could be a low frequency switching type, or even an analog type, although I'd suspect it would need considerably more transistors for that. The 7.5 HP is 22A rms, or 31A peak.

An interesting piece of history, nit very practical, since the parts used are 40 years out of date and quite hard to find these days. I was using similar parts in high power audio amplifiers back then, as were most makers of audio amps, although a guy I know started a company making switching amps for audio in that time period.

For actual use, I suggest a nice new VFD........
 
The TO3 power transistors will undoubtedly be 3 'H Bridges' used to chop the DC that is derived from your single phase supply and rectifiers applied to that bank pf large electrolytic condensers. This chopping provides the three phases of the output, one phase per H bridge

The boards jutting out at 90 degrees from the H bridges will be level shifters and possible optical isolators to bring the voltage down to be compatible with the output of the logic circuit in the lid.

Photo is too small for detail to show but I suspect that the logic board is based on TTL or CMOS standard ic's of the time producing the varying frequency drive to the power circuit. Looks fairly straight forwards to work on.

Close up focused pictures of that board would help
 
Looks a little on the light side for 7.5 HP to me. What size are the 2 fuses in holders on the input side? That may give a clue.
 
Photo is too small for detail to show but I suspect that the logic board is based on TTL or CMOS standard ic's of the time producing the varying frequency drive to the power circuit. Looks fairly straight forwards to work on.

Close up focused pictures of that board would help

Thanks thats a great explanation of what the different boards do. I took several close up's of the top and bottom boards and the power components but they seem to loose resolution when I attach them.


UNADJUSTEDNONRAW_thumb_d88b.jpgUNADJUSTEDNONRAW_thumb_d88c.jpgUNADJUSTEDNONRAW_thumb_d890.jpgUNADJUSTEDNONRAW_thumb_d88e.jpgUNADJUSTEDNONRAW_thumb_d886.jpg
 
Looks a little on the light side for 7.5 HP to me. What size are the 2 fuses in holders on the input side? That may give a clue.

I thought the same thing too when I first saw it. Looks like the fuses are UND Lab 45 amp buss fuses and I think the model numbers are DG 06 23, and DF86-95. Here are some shots of the other power components in case they have any clues about the capacity.




UNADJUSTEDNONRAW_thumb_d8a2.jpgUNADJUSTEDNONRAW_thumb_d89b.jpgUNADJUSTEDNONRAW_thumb_d897.jpgUNADJUSTEDNONRAW_thumb_d894.jpgUNADJUSTEDNONRAW_thumb_d883.jpg
 
the visible TO-3 transistors are just the drivers for.. probably darlington transistors.. probably the size of a pack of cigarettes.. bolted to the heat sink on the back.
 
the visible TO-3 transistors are just the drivers for.. probably darlington transistors.. probably the size of a pack of cigarettes.. bolted to the heat sink on the back.



Maybe. The pack type units were not really that available back then. They might have been custom made for a line of product.

There are 6 "composite transistors" visible, which may be drivers, or may be the actual output devices. For switching, there would not be so many needed, as the effects of "second breakdown" would be less of a problem, the high voltage and current condition could be very short. That many transistors could be entirely adequate, depending on the current capability of the individual parts. Ratings of 15A per device were not that rare, the usual limits were the second breakdown limits, not the actual max current.

For an analog version, which did exist at lower power (I have used them), there would have had to be several times more transistors at that sort of voltage.

Each "block" of 4 devices seems to be a single "composite transistor" unit, with a common collector voltage, the cases are directly bolted to what appears to be a heavy bracket that likely is in turn bolted to the heatsink, with an insulator between it and the sink, since the bracket is "hot with" the collector voltage. I designed a few amplifiers that way back then, it let the transistors be on a PC board, and still connect effectively to a big heatsink.

Construction became more simple once the plastic case parts like TO247 and similar became popular.
 
looks like there are 6 identical power modules or 2 per phase, with 3 power transistors per module, so one could assume 6 transistors per phase, or maybe 3 positive and 3 negative? At any rate my experience TO3 have at least 7.5 amps per, so 22 rms is quite believable.

For the money, why not, just know once it goes there is no fixing it. It may last another 20 years or 20 minutes.


You could replace capacitors and transistors, transistors are probably matched sets, finding those would be fun
 
How long had it been sitting around without being powered? If you don't know, capacitors go bad if not kept powered up, a process called "deforming". If un-powered for more than a year, you will need to do a "reforming procedure" using a variable voltage source, something not easily found in nature. If you don't, just the simple act of applying power to them can cause them to fail almost immediately.

Even then, capacitor life is typically considered to be 7-10 years, anything beyond 15 years is considered an outlier. Based on orange the color scheme, that looks to be an old "Lovejoy" drive. Lovejoy is a Power Transmission products company (couplings, clutches, etc., now part of Timken) but for a short while in the eary 80s they made their own VFDs and I seem to recall that facility was somewhere in PA. But they got out of the electronic drives business about 30 years ago, so that would put your capacitors squarely in the "exceeded their expected lifespan" category.
 
Sorry, but while caps do die and when something reaches 25 years old it is certainly the first suspect, but 7-10 years is not realistic. 15 years an outlier? Get serious

Hell, you may not have finished paying off your VMC and your caps are all bad?

Not a chance

Unpowered for a year?

You turn it on without even thinking twice about it, unless it says NASA on the side and all the designers are dead
 
I have equipment from the 1950s that has no capacitor problems with electrolytics, although some other types were leaky and had to be replaced.

To be fair, those do not "work" the capacitors as hard.

With a VFD, where the capacitors may have to take a good deal of ripple current, there is a heat induced lifetime shortening that depends on the design parameters. About 10C shortens life by half or more for most components, and for electrolytics it may be the "more".

There are usually formulas in the application data for manufacturers which will give estimated lifetime with any particular conditions of use. If the designers of the equipment have opted to beat hell out of the capacitors, a lifetime of 10 years might not be so surprising.

BUT, that is "on" time at full load, or whatever was assumed. If the actual "on" time at the conditions assumed is less, then the lifetime in years of age may be very significantly longer than what the lifetime in hours of use might imply.
 
It is good practice when returning electronics to use after a long idle period to bring up the input mains slowly over a period to give the capacitors time to reform.

I brought an ancient Chipturn CNC lathe back to life that had humongous big electrolytic caps by isolating them, and charging them from a lab supply in current limit mode at only a few milliampere over night . It had been powered of for nearly 20 years
 
It is good practice when returning electronics to use after a long idle period to bring up the input mains slowly over a period to give the capacitors time to reform.


Perhaps if you are an EE and understand the ramifications of providing completely the wrong voltage for a particular piece of equipment, but to my mind you risk blowing out things as much as you would with bad caps
 
It is good practice when returning electronics to use after a long idle period to bring up the input mains slowly over a period to give the capacitors time to reform.

I completely agree with this when you are talking about older electronics, especially vacuum tube based consumer electronics. However newer equipment starting around the mid 1970s is more likely to have SMPS or switched mode power supplies. These power supplies make the concept of bringing up the mains slowly ineffective, and risks major damage to the equipment. Reason is that the supply will try to output the full voltages to other components at an indeterminate point as you bring up the mains, while the input voltage is still much less than what it was designed for. In general, VFDs fit into this category, because they are themselves similar to switched mode power supplies.
 
Reference to one of the six small boards. What are the device numbers on the three TO-3's and the TO-66 parts.
I’ll try to add some close ups later but it looks like the numbers on the TO-3’s seem to be GE 5062 8152 and the TO-66’s look like RCA 2N6079.
 
The 2N6079 was easy to lookup. The GE part number??? The "8152" might be the date code, 1981 week 52.
The device(s) underneath the half-bridge drive boards should be looked at. I would remove them from heat sinks
and apply new thermal compound with new insulator sheets.
 








 
Back
Top